McMaster University, Hamilton, ON, Canada.
Research Institute of the McGill University Health Centre, Montreal, QC, Canada.
Arch Osteoporos. 2019 May 16;14(1):53. doi: 10.1007/s11657-019-0598-x.
Using data from the Canadian Multicentre Osteoporosis Study, several risk factors predictive of imminent (2-year) risk of low-trauma non-vertebral fracture among high-risk women were identified, including history of falls, history of low-trauma fracture, poorer physical function, and lower T score. Careful consideration should be given to targeting this population for therapy.
Fracture risk assessment has focused on a long-term horizon and populations with a broad risk range. For elderly women with osteoporosis or low bone mass, or a history of fragility fractures ("high-risk women"), risk prediction over a shorter horizon may have greater clinical relevance.
A repeated-observations design and data from the Canadian Multicentre Osteoporosis Study were employed. Study population comprised women aged ≥ 65 years with T score (total hip, femoral neck, spine) ≤ - 1.0 or prior fracture. Hazard ratios (HR) for predictors of low-trauma non-vertebral fracture during 2-year follow-up were estimated using multivariable shared frailty model.
The study population included 3228 women who contributed 5004 observations; 4.8% experienced low-trauma non-vertebral fracture during the 2-year follow-up. In bivariate analyses, important risk factors included age, back pain, history of falls, history of low-trauma fracture, physical function, health status, and total hip T score. In multivariable analyses, only four independent predictors were identified: falls in past 12 months (≥ 2 falls: HR = 1.9; 1 fall: HR = 1.5), low-trauma fracture in past 12 months (≥ 1 fracture: HR = 1.7), SF-36 physical component summary score (≤ 42.0: HR = 1.6), and total hip T score (≤ - 3.5: HR = 3.7; > - 3.5 to ≤ - 2.5: HR = 2.5; > - 2.5 to ≤ - 1: HR = 1.3).
Imminent risk of low-trauma non-vertebral fracture is elevated among high-risk women with a history of falls or low-trauma fracture, poorer physical function, and lower T score. Careful consideration should be given to identifying and targeting this population for therapy.
使用加拿大多中心骨质疏松症研究的数据，确定了几个预测高危女性近期（2 年）低创伤性非椎体骨折风险的因素，包括既往跌倒史、既往低创伤性骨折史、身体功能较差和 T 评分较低。应慎重考虑针对该人群进行治疗。
骨折风险评估一直侧重于长期和广泛风险范围的人群。对于患有骨质疏松症或低骨量的老年女性，或有脆性骨折史（“高危女性”），短期风险预测可能具有更大的临床相关性。
采用重复观测设计和加拿大多中心骨质疏松症研究的数据。研究人群包括年龄≥65 岁、T 评分（全髋、股骨颈、脊柱）≤-1.0 或既往骨折的女性。使用多变量共享脆弱性模型估计 2 年随访期间低创伤性非椎体骨折预测因子的风险比（HR）。
该研究人群包括 3228 名女性，共贡献了 5004 次观察；4.8%的女性在 2 年随访期间发生低创伤性非椎体骨折。在单变量分析中，重要的危险因素包括年龄、背痛、既往跌倒史、既往低创伤性骨折史、身体功能、健康状况和全髋 T 评分。多变量分析中仅确定了四个独立的预测因子：过去 12 个月内跌倒（≥2 次跌倒：HR=1.9；1 次跌倒：HR=1.5）、过去 12 个月内低创伤性骨折（≥1 次骨折：HR=1.7）、SF-36 身体成分综合评分（≤42.0：HR=1.6）和全髋 T 评分（≤-3.5：HR=3.7；>-3.5 至≤-2.5：HR=2.5；>-2.5 至≤-1：HR=1.3）。
有既往跌倒或低创伤性骨折史、身体功能较差和 T 评分较低的高危女性，发生低创伤性非椎体骨折的近期风险较高。应慎重考虑针对该人群进行治疗。
